YorkieTalk.com Forums - Yorkshire Terrier Community - View Single Post

tricia208 · 09-09-2013, 07:46 PM

I started reading this thread several times a few months ago and finally made myself finish it today. It is a good read with great information. I have been researching MARS and their breed recognition products for a project over the last few years. While I cannot state with any authority that I understand algorithms and the science behind the MARS testing, I do have a better understanding about interpreting their test results and the limitations of the same. On background, MARS did not develop any of the products they sell today as you probably already know. They purchased competitor companies, protocols and the control of licenses and patents that are sold today as the MARS Wisdom Panel products. They do not perform the Wisdom Panel testing. A company in Nebraska performs all the testing, stores the data and provides the test results to the customer. (I know the name of the company but they would not like me if I posted it.) That is just the way MARS conducts their business. It is not unusual or improper in the business world and should not reflect on the quality of their products.

My guess from looking at "absolutely adorable Minnie" its more than likely that she does have another breed(s) however, the breed may not be listed on her results for several reasons. The missing breed may not be in the MARS database therefore cannot be listed. Their goal is to have all AKC recognized breeds in their database but they are not there yet. In some cases, the MARS test results will state a percentage of "Unknown Breed" to note the breed not in their database along with the percentage of known breed(s). Other MARS test results will not include "Unknown Breeds" even though DNA is present. The percentage may be too small to list. Another reason that may explain Minnie's results is the accuracy of the MARS products are 90% at best. A few years ago MARS stated on their website that their accuracy rate was 84%. They have since removed that information. I was able to find some quotes in published articles from their staff claiming 90% accuracy. About a year ago MARS was offered a large project to test hundreds of Norwich terriers. We will save the story behind the Norwich terrier for another day. MARS turned down the project stating the reason was "the MARS test was fallible". This statement and exact wording was made by the MARS representative at the AKC Board of Directors meeting. Fallible was his exact word. He then mentioned that one out of ten tests results would be inaccurate. So it is easy to see why Minnie's results may or may not have all breeds listed and I just listed a few of the reasons.

In the next 5 years MARS will be the only company providing breed recognition and ancestry products to the public. They are buying and controlling every company that offers a breed identification or ancestry product. MARS has staff with the ability to develop these products but they decided to buy all their competitors probably because it is faster and cheaper in the long run.

Here are two fairly recent articles about the MARS company and their test results that you may find interesting.

Dog breed tests: Do they really work? | MNN - Mother Nature Network

Dog breed genetic tests put to the test - VIN

I believe that MARS offers a good service and product but I also believe it has its limitations. I have also witnessed individuals with an agenda or wanting certain test results therefore misleading the public about the capability of the MARS products. Personally, I would only recommend a MARS product to someone interested in acquiring a rescue or shelter dog. That was the original reason MARS went into the DNA Breed Recognition business and it is the best product they offer today. That is mainly due to the customer understanding the results are "fallible" and have limitations. I also understand that some people are just curious about their dog's ancestry and are willing to spend the money. As long as those individuals are aware the product in its present state has limitations then I would recommend their products.

It looks like Minnie is the real winner in all of this discussion. It really does not matter what breed Minnie is other than the OP would have posted on a different forum and we would have all lost out on the dialog, education and fellowship.

09-09-2013, 07:46 PM	#622
tricia208 Senior Yorkie Talker Join Date: Feb 2012 Location: Ocean Beach, CA Posts: 109	I started reading this thread several times a few months ago and finally made myself finish it today. It is a good read with great information. I have been researching MARS and their breed recognition products for a project over the last few years. While I cannot state with any authority that I understand algorithms and the science behind the MARS testing, I do have a better understanding about interpreting their test results and the limitations of the same. On background, MARS did not develop any of the products they sell today as you probably already know. They purchased competitor companies, protocols and the control of licenses and patents that are sold today as the MARS Wisdom Panel products. They do not perform the Wisdom Panel testing. A company in Nebraska performs all the testing, stores the data and provides the test results to the customer. (I know the name of the company but they would not like me if I posted it.) That is just the way MARS conducts their business. It is not unusual or improper in the business world and should not reflect on the quality of their products. My guess from looking at "absolutely adorable Minnie" its more than likely that she does have another breed(s) however, the breed may not be listed on her results for several reasons. The missing breed may not be in the MARS database therefore cannot be listed. Their goal is to have all AKC recognized breeds in their database but they are not there yet. In some cases, the MARS test results will state a percentage of "Unknown Breed" to note the breed not in their database along with the percentage of known breed(s). Other MARS test results will not include "Unknown Breeds" even though DNA is present. The percentage may be too small to list. Another reason that may explain Minnie's results is the accuracy of the MARS products are 90% at best. A few years ago MARS stated on their website that their accuracy rate was 84%. They have since removed that information. I was able to find some quotes in published articles from their staff claiming 90% accuracy. About a year ago MARS was offered a large project to test hundreds of Norwich terriers. We will save the story behind the Norwich terrier for another day. MARS turned down the project stating the reason was "the MARS test was fallible". This statement and exact wording was made by the MARS representative at the AKC Board of Directors meeting. Fallible was his exact word. He then mentioned that one out of ten tests results would be inaccurate. So it is easy to see why Minnie's results may or may not have all breeds listed and I just listed a few of the reasons. In the next 5 years MARS will be the only company providing breed recognition and ancestry products to the public. They are buying and controlling every company that offers a breed identification or ancestry product. MARS has staff with the ability to develop these products but they decided to buy all their competitors probably because it is faster and cheaper in the long run. Here are two fairly recent articles about the MARS company and their test results that you may find interesting. Dog breed tests: Do they really work? \| MNN - Mother Nature Network Dog breed genetic tests put to the test - VIN I believe that MARS offers a good service and product but I also believe it has its limitations. I have also witnessed individuals with an agenda or wanting certain test results therefore misleading the public about the capability of the MARS products. Personally, I would only recommend a MARS product to someone interested in acquiring a rescue or shelter dog. That was the original reason MARS went into the DNA Breed Recognition business and it is the best product they offer today. That is mainly due to the customer understanding the results are "fallible" and have limitations. I also understand that some people are just curious about their dog's ancestry and are willing to spend the money. As long as those individuals are aware the product in its present state has limitations then I would recommend their products. It looks like Minnie is the real winner in all of this discussion. It really does not matter what breed Minnie is other than the OP would have posted on a different forum and we would have all lost out on the dialog, education and fellowship.